Primary Sources for Geographers

Martha B Sharma & Gary S Elbow. Using Internet Primary Sources to Teach Critical Thinking Skills in Geography. Greenwood Press, 2000.

Geography is related to many other fields of study. These fields include sciences such as geology, meteorology, soil science, biology and hydrology, which most often are part of the study of physical geography. Human geographers use data and concepts from social sciences such as economics, anthropology, linguistics, political science, psychology, and sociology as well as material from humanities disciplines such as history, languages, and literature. Because of its diversity geography depends on an extremely wide array of data sources to describe different parts of the world, to identify relationships among different parts of Earth, and to explain the processes that account for where things are and why they occur as they do. Data sources are often placed in the following categories: primary, original, or secondary.

Primary Sources, Original Sources, and Secondary Sources

Primary data sources used by geographers (Table 2.1) may be written in narrative form as notes, diaries, or other accounts of important events or processes, they may be handdrawn field maps or published base maps, or they may be quantitative information such as census counts or temperature and precipitation data. Audio or video recordings, photographs, and digitized data from remote sensing can also be primary data. The thing that makes the data useful for geographers is that it tells us something about Earth and its human population. Primary sources need not be original, but they should have been carefully compiled or prepared from original information so they have a high probability of being accurate and free of bias. Original sources are materials that come directly from the individual (or instrument) that collected them. These include ships’ logs, scientific field notes, first-person accounts of travel or investigation, original photographs (ground or aerial), satellite images and digital data, and original data that is collected on the ground by researchers and/or instruments such as the census-takers’ reports from a national census or climate data gathered at a weather station, for example. Original sources are often kept in archives, special collection libraries, or other secure places. Access to originals is often restricted, but microfilm copies, photocopies, or printed versions may be available. Original sources are primary sources but not all primary sources are original sources as we shall see below.

Table 2.1

Primary Data Sources Used by Geographers

  • letters, diaries, journals, field notes, and other written accounts that describe landscapes, native peoples, economic activities, or other information that is of potential interest for geographers
  • data collected in the field by the investigator, such as field maps, interviews, surveys, photographs, or digital images
  • data on human conditions such as censuses, economic production data, literacy rates and levels of education, health conditions and/or care, and type and quality of housing, which provide information on demographic, social, economic, or other human characteristics of a region, country, or group of people
  • data on physical conditions such as temperature, precipitation, wind, cloud cover, soil characteristics and type, elevation, slope, drainage, vegetation cover, and other information relating to the physical conditions of the area being studied
  • maps of certain types such as field maps, topographic maps (for place names, landforms, stream patterns, transport systems, etc.), and photo maps
  • photographs, digital images, motion pictures, videos, and audio tapes (if they contain relevant information)
  • aerial photographs, satellite images, and other remotely sensed data in photographic or digital form

Secondary sources, which are not a focus of this book, include secondhand reports such as might be contained in biographies, condensed or abstracted travel accounts, general histories or geographies, data that have been condensed or statistically manipulated, or any other source that has been modified in some substantive manner or that condenses or summarizes from primary or original sources. The assumption is that such data, while useful, may contain errors or interpretations that have been added in the course of their publication or presentation. Written documents may have been abridged, taken out of context, or have had errors introduced in the process of reproduction or quotation. None of these limitations mean that information from secondary sources is not valid or potentially valuable, but it does warn the user to be aware of possible errors, omissions, or biases that might be introduced beyond those that were inherent in the primary source from which they were derived.

The following example illustrates the difference between primary, original, and secondary geographic sources. Christopher Columbus kept a diary of his first voyage of discovery to the New World in 1492 (diary excerpts are on the Internet at the Medieval Sourcebook site: Columbus presented the original of this invaluable document to the King and Queen of Spain when he returned from his epic journey and a hand-written copy was made for him. The diary is unquestionably an original source and a primary source. If the copy were perfectly made, it could be considered a primary source as well, but it would be necessary to compare the two documents to determine if they contain identical information because the person who made the copy might have made errors in transcription.

Unfortunately, both the original Columbus diary and the copy have been lost for centuries. We now depend on an abstract, or summary, of the diary made by the Dominican friar Bartolome de las Casas probably about forty years after Columbus made his voyage. No one knows for sure how much of the original diary Las Casas may have left out, what he summarized or took out of context, and what he copied verbatim. Also left open is the possibility of transcription errors that may have been made as Las Casas copied from the Columbus diary. The Las Casas version of Columbus’s diary is clearly not an original source. If the original were still in existence, the Las Casas copy would be a secondary source and probably of little interest for historians, geographers, and other researchers. However, since the Las Casas transcription is the only contemporary copy of the Columbus diary that remains in existence, it may be considered a primary source document for the Columbus voyage, even though it is clearly far less than the original (Dunn and Kelley 1989).

Uses and Reliability of Primary Sources in Geography

Geographers use primary sources to learn about Earth. Primary data about the physical world such as rainfall and temperature records, distribution of earthquakes and volcanic activity, records of vegetation types, flood records, or soil characteristics all contribute to understanding the natural processes that shape the different parts of Earth. Data on population distribution, type of economy, quality of life, health status, education, and a wide range of other social and economic indicators are essential for understanding the different ways in which humans use natural resources. They also give clues to how well or poorly they live. Historical records of all sorts, from surveyor’s reports of vegetation and other patterns of landscape to aerial photographs taken decades ago, help geographers and other scientists discover how Earth is changing and why (see the USGS Earthshots: Satellite Images of Environmental Change site at for satellite images that illustrate environmental change over the past twenty years).

Generally, information from primary sources is preferred for all of the purposes noted above. That is because primary sources are less likely to contain errors than data that has been processed in some way and because the primary sources contain basic data that can be interpreted or, in the case of numerical data, manipulated statistically to reveal useful and interesting trends or details. Photographs and maps that are simplified or generalized contain less information than the primary sources from which they were derived. In the case of historical materials, the primary sources may provide the earliest recorded impressions of what a place looks like, or they may provide benchmark data that can be used to determine what kind of change has taken place or is taking place in a particular area. Secondary sources may provide the same information but they are considered to be less reliable because the information has been modified by a second party, as in the case of the Las Casas abstract of the Columbus diary.

As noted above, primary sources are usually considered to be better than secondary sources for most purposes. Nevertheless, they do present some problems for the user. Perhaps the most obvious problem is the reliability of first-person accounts. Returning to the example of Columbus, when he arrived in the Caribbean in 1492 he literally did not know where he was. He was convinced that he had sailed to Asia and many of his interpretations were shaped by that error. First person accounts may be interesting, but they are not necessarily reliable. Students who use such accounts must read with caution and look for inconsistencies in narration and reports of unlikely events that are clues to possible errors of fact or interpretation.

With respect to statistical sources, most of us are familiar with the computer user’s warning: “Garbage in, garbage out.” No matter how sophisticated the analysis or how powerful the computer, the results of analysis of numerical data are only as good as the information from which the analysis was produced. Data that contains bias, is incomplete, was recorded incorrectly, or is based on misleading or confusingly worded questions is reduced in value or useless as a tool for interpreting processes of change on Earth. For these reasons, it is important to know the conditions under which the data was collected, including the definitions of certain terms used in censuses, instructions to the data collectors and, if a sample survey was used, how the samples were constructed.

Even data collected by scientific instruments may present problems. Rainfall records may be increased or reduced depending on where the rainfall gauge is located. A thermometer located near an asphalt street or parking lot will register higher and lower extremes than one in a grassy area, and a sunny location will get much hotter than one in the shade. Los Angeles presents an example of how location can affect weather and climate data. The airport, where official climate data is recorded, is located in west Los Angeles, close to the coast. Cool ocean breezes keep average summer temperatures at the airport about 10°F cooler than locations even a few blocks farther east and more distant from the coast. Most people in Los Angeles live inland away from the coast. For them temperatures recorded at the airport have little or no relevance, and the same might be said for someone who was planning a visit to the city and who wanted to know what kind of weather to prepare for (this bit of information may be tested by going to a primary source, the National Atmospheric and Oceanic Administration at, and checking temperature data for different locations in the greater Los Angeles urban area).

One would expect that photographs and maps, which seem to represent the world “as it is,” would be free from bias and error. However, no map can include everything that appears in the area of coverage. The cartographer must decide what elements are most important for the purpose of the map and which items may be left out. Thematic maps show specific distributions, for example climate, land use, or some other phenomenon that may be of interest. They usually omit any information that is not directly related to the theme of the map or necessary to show distributions (such as continents, oceans, national boundaries, and the geographic grid). For these reasons, thematic maps are generally considered to be secondary sources.

Base maps such as the topographic maps produced by government agencies in the United States and other countries (see Chapter 3 for a short list of these agencies) include a wide range of cultural data such as roads, railroads, buildings, airports, and key landmarks such as water towers, reservoirs and dams, beacons and lighthouses, bridges, schools, churches, and cemeteries. Physical information includes lakes, rivers and streams, landforms and elevations (shown by contour lines), vegetation cover, and many other features. However, a great deal of information is omitted or generalized in order to keep the map from becoming too crowded or busy. It is easy to see what has been included in a map, but the only way to know what has been left off is actually to visit the map site or to look at aerial photographs or satellite images of the area and compare them with the map. The choice of what information to include and omit from a map affects its usefulness and the user should be aware in a general way of what is included and what has been left off a map to obtain the greatest value from it.

In a technical sense topographic maps are secondary sources because there is a selection process for data to be included on the map. However, many geographers consider such maps to be primary sources because they provide faithful portrayals of landforms and other physical characteristics along with an assortment of cultural features that makes up the base data of the map. Old editions of topographic maps (some early European topographic maps are now over two centuries old) are invaluable historic records of past landscapes. However, as noted above, the user must be aware that some details of the landscape were omitted from the map.

Even aerial photographs and satellite images present problems. Conditions may change rapidly and the image, unless it is being used for historical comparison, should be recent. Whatever the image, if it does not carry an indication of the location it covers and a date, it is virtually worthless. Location is fundamental for geographers and the photograph should carry an identifying place name as well as approximate geographical coordinates (latitude and longitude) so the subject can easily be found on a map. The year of the photograph is a benchmark that allows the user to place the photograph in its historic context and one should know the time of year when an image was generated because certain patterns occur seasonally. Certain features are easier to see in mid-latitude winter, for example, when deciduous trees are bare of leaves (the same principle applies in seasonal tropical areas, where trees often lose their leaves in the dry season). Some objects may be difficult to identify on a normal photograph, such as a house tucked away in the forest. On the other hand, infra-red films and sensors can register the heat generated by a warm building in winter even if it is not easily seen on film that is sensitive to visible light. Scale can also be a problem with photographs, which undergo slight distortion between the center and the edges because of the curvature of the lens and the changing perspective from the point where the photograph is centered. There are ways of compensating for or correcting these problems but one must know they exist before they may be dealt with.

Finally, there is the issue of geographic scale and the level of aggregation of data. In a strict sense, the primary data from a census or survey is the raw information contained in the census questionnaires or survey documents. However, these sources may be confidential. For example, the raw data from the Untied States census (the actual census takers’ records) are not available to the public until eighty years after the census was made. Even if the information were available to the public, for most purposes the time required for assembling it into a usable form would be prohibitive. Thus, researchers generally rely on statistical compilations of the original data. The choice of how this information is aggregated will determine the scale at which the researcher can work. If census data are available only at the state level, for example, it will be impossible to see patterns of population distribution within a state. Data by census tract may be very useful if one wants to see where different socioeconomic groups live within a city or metropolitan area but they would be difficult to handle if one were interested in state-level distributions. County data might be more useful for that scale of analysis.

Primary Sources and Scientific Data Collection for Geography

Data sources have changed over time as technology has progressed. Two hundred years ago, Lewis and Clark made sketch maps and drawings and kept hand-written diaries and notebooks to record the scientific discoveries of their famous expedition to the Pacific Coast. They collected and dried plant specimens and caught wildlife, some of which was returned live to Washington, D.C. However, many of the biological specimens were kept as skeletons and dried skins that were carried along with the expedition until it returned to the United States. Latitude was calculated using a sextant and longitude was determined with a chronometer. Lewis and Clark made careful observations of wildlife, vegetation, and landforms, and plotted the courses of rivers they followed. These observations were made in accord with the instructions the explorers received from President Thomas Jefferson, which were to collect as much information about the land and native people as they could and to record information from which an accurate map of their route could be made (see the Rojomo site at∼rojomo/landc.htm?12,44 for the instructions that were given to Lewis and Clark by President Jefferson).

Today an expedition to Antarctica, for example, would carry an array of sophisticated instruments to measure every conceivable natural characteristic of the area and use modern global positioning systems (GPS) to determine exact geographic locations and surface elevations. Any biological specimens that were collected would be sent back live, if possible, for classification and analysis. If this were not possible, specimens would be frozen for later shipment to laboratories for examination using high powered microscopes and chemical analyses.

In the areas of social analysis Lewis and Clark lived with some native Americans and visited others. They reported what they observed of native habits, commenting on agricultural practices, hunting techniques, food preparation, belief systems, and dozens of other aspects of their life, and they sent examples of their weapons, tools, and other artifacts to Washington along with the biological specimens noted above.

In modern times social and economic data is still sometimes collected in the field by researchers, but it is more common to use surveys, censuses, or official reports from government agencies to learn about how people live. Many times the data collection procedures are standardized by international organizations such as the United Nations in the hope that information will be comparable for all countries that participate. However, there is no mechanism to enforce such standardized procedures and international comparisons may be affected by differences in the standards used to collect and/or analyze it or by the currency of the information made available. The World Bank web site ( contains a discussion of some of these data problems on its page titled “Primary Data Documentation,” which can be downloaded as a PDF file.

Remote sensing, the collection of data from cameras and other sensors in airplanes satellites, and other remotely located platforms, also is used to obtain data on social and economic characteristics of human population on Earth. Remote sensing is especially useful for observing the spatial distributions of human activities on Earth, which is a special interest of geographers. The “What’s New In Remote Sensing” page ( provides links to many interesting sites dealing with remote sensing. One of the links at this site, the European Space Agency, offers a series of classroom exercises for students to learn about remote sensing and its applications (

Analytical techniques also have become much more sophisticated as knowledge about the world improves, as the amount of data available increases, and as computers make it possible to process that data easily and rapidly. Geographic information systems (GIS) allow geographers to overlay vast amounts of data obtained by remote sensing and to compare it, providing new insights into how we humans use and modify Earth. There are many private companies that offer GIS packages and most university geography programs have GIS courses. To access information on these resources try the page ( on GIS. This site provides links to other sites and also provides a brief introduction to GIS technology.

Early geographers offered speculative explanations of change on Earth that were based on legend, superstition, or intuition. Modern geographers publish rigorous scientific studies that are derived from analysis of large amounts of data. The data come from a multitude of sources, including those mentioned above, and the theories developed from that data are subject to continual testing and reevaluation in the light of new information. New data may lead to development of alternative hypotheses that are perceived to offer more satisfactory explanations of natural and/or social processes on Earth. This continuous testing and reevaluation of explanations for geographic phenomena are part of the scientific method. The scientific method, in turn, is closely linked with the strategies of critical thinking: comparing and contrasting, sequencing, classifying, identifying cause and effect, problem-solving, and decision-making. (There are many web sites dealing with scientific method and hypothesis building. The following site is part of “Fundamentals of Physical Geography,” an Internet-based course developed by Dr. Michael Pidwirny of Okanagan University College in British Columbia, Canada. This page discusses scientific method in the context of physical geography:

The scientific method depends on accurate data for its effectiveness. If data are absent or incorrect, the conclusions drawn from research will be questionable at best and at worst wrong. Therefore, any effective geographic investigation begins with a strategy for obtaining the most accurate data available on the subject at hand. Primary sources generally offer the best possibility for obtaining data of the quality, reliability, and replicability needed to support the rigorous testing required by the scientific method.

First-Person Accounts

Many travelers keep diaries, notes, or other informal records of the things they observe as they visit new and strange places. Geographers and other scientists generally keep fairly extensive field notes from which they later draw conclusions or extract information. Some of the records kept by Meriwether Lewis and William Clark, cited in the previous section, are field notes but the documents that are usually reproduced are the journals kept by the expedition leaders and their men (see “Along the Trail with Lewis and Clark,” a site created by students at Washington State University, for excerpts from the Lewis and Clark Expedition journals at This is because field notes are not usually kept in narrative form so they are much less interesting to read than journals which record the day to day events of travel and exploration. Some other journals and informal travel diaries available on the Internet are noted in the sections that follow.

Both informal travel diaries and explorers’ or scientists’ journals may provide useful geographic insights, but they are quite different. Geographers and explorers generally provide more thorough accounts, and often take pains to note details about the landscape, including plants and animals, landforms, climate, native people, and other features because they know they will have to prepare a formal report on what they have observed in their travels. Diaries and personal notes are kept for the satisfaction of the individuals who make them and they are generally less systematic in their recording of geographic information than journals.

Explorers and Geographers

Often geography is equated with exploration and some of the most compelling accounts in the history of geography, namely the diary of Columbus’s first voyage, come from explorers. Unquestionably, exploration expands our knowledge of Earth. However, most explorers are motivated by non-scientific objectives. Explorers want to forge new trails, to go where no one has gone before. Columbus was the first European to see the Caribbean. Magellan commanded the first fleet to sail around the world. Vasco da Gama was the first European to sail around Africa to India. Roald Amundsen was the first person to stand at the South Pole and Admiral Richard Byrd was the first to fly over it. In their quest to be first, explorers may also be searching for knowledge that will have military or economic value and their observations are clearly useful for geographers and other scientists. Nevertheless, explorers generally are not trained to collect data of explanatory value nor are they trained to seek explanations for what they see. Columbus, for example, believed he had found cinnamon trees in the Caribbean, but none grew there at the time of his voyage. If Columbus had known more about botany or if there had been a botanist in his crew, he probably would not have made this mistake. It is the quest that motivates explorers. Scientific objectives are secondary and often included only because they justify the exploration to fund-granting organizations and individuals.

Geographers, on the other hand, along with natural historians, geologists, anthropologists, and other students of the natural world and its people, are trained to collect data carefully, recording all of the conditions under which it is gathered, and to use these data to improve our understanding of natural and social processes. Their methods and objectives differ in important ways from those of explorers. They take copious notes, collect samples of plants, animals, rocks, soil, and whatever human artifacts may be available, and carefully document the exact location and circumstances under which they collected specimens or made observations. Geographers think comparatively, trying to relate what they observe to other situations they may have seen or read about. How are two places similar and why? How are they different and what might explain the differences? What are the important processes that shape the landscapes I am seeing? These are the questions geographers ask as they collect data. As fascinating and compelling as first-person accounts of exploration may be, the often prosaic reports from the field of scientists frequently may be more useful primary sources for understanding the world.

NPR’s “Radio Expeditions” series (, which aired in 1999, is an online audio presentation that deals with exploration and research. Each program, which lasts for about ten minutes, features an important event of exploration or scientific discovery. These programs are not primary sources in the strict sense, because they are, for the most part, second-hand reports. However, a few of the programs feature interviews with living scientists and explorers or quotations from their writings. After a class discussion of the difference between scientific research and exploration, students could evaluate which type of activity a program describes and explain why they chose the classification they did. As an alternative, students could report on what the scientist or explorer had to say about her or his work.

Diaries and Travel Narratives

Diaries and travel narratives are personal accounts generally written by non-specialists who may be keen observers of the world. The travel account of Marco Polo is one of the earliest such documents (see the Silk Road Foundation page on Marco Polo at or the Geography Site Marco Polo biographical sketch at It is still read by historians and geographers who seek knowledge of Asia as it was nearly 800 years ago; and the veracity of Polo’s account is still hotly debated (see USA Today Book Shelf for a review of a book that argues Marco Polo never went to China at More contemporary examples of personal accounts are diaries of pioneers as they traveled across the United States on the Oregon Trail or to California during the Gold Rush, reports from people who have visited foreign lands, or accounts of some natural disaster. All fall into the category of personal accounts. Many such accounts may be found on the Internet. The Library of Western Fur Trade Historical Source Documents (∼drudy/mtman/mmarch.html) contains links to many excellent examples of this genre.

For students who live in communities located near pioneer trails or in other places for which personal accounts may be available (large cities, for example) it may be possible to use historical accounts to identify elements of landscape change between the time of the report and the present. Questions that might be asked about rural areas are: How has vegetation changed? Are any of the old landscapes preserved? If farms are mentioned, are crops still the same? Are pioneer travel routes paralleled by modern highways or railroads? Are specific places that are noted in the travel accounts still identifiable on the landscape?

If students live in an urban area, the questions will involve changes in the urban landscape. What means of transportation were mentioned in the historic accounts? Do these means of transportation still exist, or have they changed? In what ways have neighborhoods been transformed? How has the city expanded? What kinds of city services were mentioned in the accounts, and have they changed? How has the ethnic composition of the city changed?

Increasingly, libraries and archives are digitizing historic materials, so travel accounts such as those described above may be available on local library or historical association web sites for many places. If such materials are not available locally, students could access more general accounts such as Lewis and Clark’s journals (available on the PBS site at or Oregon Trail pioneer accounts (available on the End of the Oregon Trail Interpretive Center site at∼eotic/).

Accounts of trips to a foreign country abound on the Internet and may be located by looking under the country name. For example, a very interesting illustrated account of a trip to Mongolia titled “Greg’s Mongolian Journal” appears at Such accounts may be geographically naïve when compared with the reports of professional explorers or scientists, but they compensate in several ways. They are written in lay person’s terms, which may make them more accessible for students. They reveal information about how people live, about how places look, and about the traveler’s impressions, which may be left out of more formal reports.

First-person reports by survivors of natural disasters are also easy to locate on the Internet. They commonly appear soon after a hurricane, earthquake, tornado, volcanic eruption, or tsunami, for example. The Miami Museum of Science hurricane site at has information for younger students that includes a family’s account of surviving Hurricane Andrew in 1992. Another good hurricane site with activities for older students (6-8/9-12) is at ( This site has excellent links to other hurricane-related web sites. First-person accounts of disasters reveal a great deal about the event itself and also may provide information on the reactions of official agencies and individuals to the crisis. For places that have the potential to suffer a particular type of natural disaster, for example, hurricanes on the Gulf or Atlantic Coast, or earthquakes in the western United States, first-person accounts may also be instructive for student disaster preparedness training. For some disasters that occur in non-English-speaking areas it may be difficult to locate survivor accounts in English. Hurricane Mitch, which impacted Central America in 1998, and the Sisani Lagoon tsunami of July 17, 1998, in the country of Papua New Guinea, are two recent examples where many newspaper accounts and other secondary sources are available but primary sources are hard to locate.

Collecting Data from the Field

Early geographers described the things they saw using their own observations or first-person accounts from others as the primary data source on which to base their descriptions. They relied on their own deductive ability to explain what they observed or read about. Typical of these early geographers were people like Charles-Marie de la Condamine and Alexander von Humboldt. De la Condamine was a French mathematician and scientist who led a seven-year expedition to South America between 1734 and 1741 to help determine the exact shape and dimensions of Earth. Von Humboldt was a German geographer who made an epic tour of the Americas between 1799 and 1804. Both of these men were part of the early effort to measure different physical characteristics of Earth and both were keen observers of the world around them. They recorded their observations and tried to develop rational explanations for what they found. The journals of geographers like de la Condamine and von Humboldt are still useful primary sources for learning what the places they visited were like at the time the reports were made. (A short biographical sketch of Alexander von Humboldt can be located on the Internet at the Geography Site at and information about Charles-Marie de la Condamine is on the History of Mathematics site at

The attempt to explain the world links modern geographers with their forebears of the Age of Enlightenment. They are still engaged in observing, explaining, and predicting. To do this they need many different kinds of data. These data come from an almost infinite variety of sources. Some geographers still collect data in the field just as de la Condamine and von Humboldt did. Students may follow this tradition by writing descriptions of the areas in which they live or through which they travel as they go to school, shopping, or on other trips through their community. Family vacations provide another possibility for students to keep a travel diary in which they record their observations as they travel to new places. When they do this, they are collecting their own primary source data. If they cannot conduct their own fieldwork, they may want to look at some of the virtual field trips on the Internet that are noted in Chapters 3 and 4.

Because modern technology has provided us with such an amazing variety of machines that help us to gather data, nowadays it is often not necessary for geographers to actually go into the field to obtain data for their studies. Geographers use information downloaded from satellites and ground sensors, or collected via censuses, questionnaires, or in a wide variety of other ways. These data sources have transformed our ability to observe and understand the processes, both natural and human induced, that are transforming Earth.

Censuses, Statistical Reports, Polls, and Surveys

Human geographers are interested in how people live and how they interact with the environment in which they live. Much of the data that human geographers use comes from censuses, surveys, and statistical reports of one sort or another. A census is a periodic count of the population usually conducted by a national government. Censuses have been around for millennia but the first decennial (every decade) census was conducted in the United States in 1790. The United States has conducted a census every ten years since then in order to determine the distribution of population to insure equality of voting in accord with the Constitution. Addresses for United States census Internet sites are listed below and in Chapter 4.

Most censuses concentrate on counting noses—finding out how many people live in a place. However, they also collect information on a wide variety of other characteristics of population. Most censuses record data on age, gender, and marital/family status (is the head of household married, divorced, single; how many people live in a residence and what is their relationship to each other). The United States census also collects information on quality of housing, employment status and occupation as well as race and/or ethnic affiliation. In the United States, the population census is now conducted by mail, and census takers will visit only households from which a mail questionnaire is not returned. Other information may be collected via a random sample that is designed to obtain data from a representative group rather than the entire population of the United States.

Most other countries also conduct regular censuses but the period between censuses, the type of data collected, the method of collection, and the consistency of data collected from census to census all vary greatly. (See Chapter 3 for information on the Mexican census web site and Chapter 4 for the German and Indian censuses.) Some countries do not collect data on ethnic affiliation, others may not report certain kinds of economic information, and still others may add or delete certain information on health or educational status. All of this makes it difficult to compare censuses conducted by different national governments.

Another important issue with respect to censuses is accuracy. The United States, which has the longest continuous record of census taking and which uses very sophisticated techniques to insure the accuracy of its census, admits to a five percent error rate. To make matters worse, the error rate varies spatially and with respect to ethnic minorities. The census tends to be more complete in rural and suburban areas where middle-class families live and less accurate in inner cities and some rural areas where there may be people living in irregular circumstances. African Americans and Hispanics are more likely to be missed by the census than Caucasians. (Papers discussing census undercounts with special reference to ethnic minorities and recent immigrants are available on the Census web site at by clicking on estimates in the “people” category. Actual data comparing “official” and “estimated” results from the 1990 census are also available at this site at the Census 2000 prompt.)

Other countries, especially those undergoing civil unrest or suffering from other kinds of problems, are likely to have high rates of error. Usually the errors are undercounts, resulting from people being missed. Occasionally individuals may be assigned to an incorrect ethnic group or otherwise placed in a wrong category. Those who regularly use census data learn to spot discrepancies that may reveal an error, but some are subtle and may persist for years before anyone finds them.

Most national governments collect a variety of socio-economic data in addition to the demographic information that is the main focus of the national census. These data may provide information on literacy and education, health facilities and health status, sanitation conditions, housing, distribution of employment across sectors of the economy, economic activity by sector, imports and exports, and many other social and economic indicators. Some of these data are collected as part of national censuses, but much of it will be gathered from other sources. Often such data is summarized annually making it more current than census data.

The extent to which this data can be considered a primary source is debatable. Data are usually released in aggregated form, for the entire country or by state, province, or other major internal division. On the other hand, the published data are the least refined source that is generally available, so some would consider such information to constitute a primary source. Many countries have Internet sites on which such social and economic data may be accessed. It is also available from some international organizations such as the World Bank (

Surveys and polls differ from censuses in several important ways. Usually they focus on a relatively narrow range of topics and they use a sample group rather than covering an entire population as a census does. Also, many polls tend to focus on topics related to marketing, politics, or economic issues and may be of marginal interest for geographers. Nevertheless, some poll data may be tabulated by region of the country and the differences among regions may reveal interesting clues to social or economic differences within the country. Some polls compare attitudes among different ethnic groups living in the United States, which likewise, is of interest to geographers. Another important aspect of surveys and polls is that they are often conducted by private organizations. The data they collect are geared toward the specific needs of the group for whom the survey is conducted and the data may not be released to the public except in summary form (i.e., the primary data are not available). Clearly, this limits the utility of polls and surveys. Still, a wide variety of poll data is available from certain web sites such as the Pew Center for the People and the Press ( and the University of Michigan Inter-university Consortium for Political and Social Research ( Not all of these polls deal with topics of interest to geographers, but some do.

Sources of Data on Environmental Conditions

There are many different sources of information on environmental conditions on the Internet. Temperature and precipitation data by month, as indicated earlier in the case of Los Angeles, are available for all major cities of the United States. Data for many other parts of the world may be accessed at the World Climate site ( or for some countries through sites maintained by their meteorological service.

Other data on physical conditions are not as readily available on the Internet. Cloud cover, sea surface temperature, and other climate-related data are relayed to Earth from satellites on a regular basis and may be accessed at sites such as NOAA (, World Meteorological Organization (, and the Lamont-Dougherty Geophysical Laboratory of Columbia University ( Soil types and characteristics are problematic because the data are complex and the maps large scale, but they can be accessed for locations in the United States at the Department of Agriculture’s Natural Resources Conservation System site ( A learning module for lower grade students on soil development is also available at this site, as is additional data on natural resources of the United States. Changes in vegetation cover can be observed from satellite images and other remote sensing sources via the United States Geological Survey at its Earthshots site that was previously noted.


Sites such as the Perry-Castanñeda Library of the University of Texas ( and the Library of Congress ( have reproduced historic maps. However, the small size and relatively poor quality of reproduction of these maps limits their utility. Historic maps are also reproduced in web sites featuring historic events such as the Lewis and Clark route, cited earlier in this chapter.

Topographic maps are analyzed in some specific sites such as Topozone ( They suffer from some of the same problems of scale and quality of reproduction as the historic maps. The United States Geological Survey maintains the “Finding Your Way with Map and Compass” site (, designed to teach students to use topographic maps to locate themselves.

Several web sites produce local street and road maps for virtually any place in the United States as well as for some foreign areas. These maps can be very useful for locating specific addresses, determining street patterns, and similar activities. For the United States, the Census Bureau’s TIGER data base is useful (, and private sources such as MapsOnUs (http://www., MapQuest (, and MapBlast ( also feature large-scale street maps.

Ground-Level Photographs

Visual images are a strong point of the Internet. Web sites abound with interesting illustrations that show places in the United States and elsewhere in the world. These illustrations range from tourist photographs to significant historical photographs and drawings. “Greg’s Mongolian Journal,” cited above, contains photographs illustrating contemporary conditions in Mongolia, one of the world’s most isolated countries. Tourist-oriented sites exhibiting professional and/or amateur photography exist for nearly every country in the world. The possibilities for developing exciting and challenging geography exercises are limited only by the imagination of the teacher. The Electronic Field Trip of Glacier National Park ( has many photographs of cultural and physical features of this famous national park. An interesting critical thinking exercise using this site could ask students to prepare a diary describing their trip and commenting on each photograph.

Aerial Photographs and Other Remotely Sensed Data Sources

Aerial photographs date from 1858, when a French photographer used a balloon as a platform to make photographs of Paris from above Earth’s surface. The first documented aerial photographs in the United States were made of Boston in 1860. Shortly thereafter, cameras borne in balloons were used to monitor troop movements and other military activity during the Civil War. Modern satellite imagery has its origin in these early experiments with photography from a unique bird’s eye perspective. For decades aerial photographs were made from airplanes on black and white film and variations in gray tones were used to interpret the resulting images. Today, in addition to images in the visible spectrum (color), a wide range of other radiation wave-lengths is used to derive images of Earth, other planets in the solar system, and distant stars, nebulae, galaxies, and other astronomical features. On Earth, infrared sensors can detect heat loss from buildings, motorized military equipment such as tanks and armored personnel carriers, and surfaces of different temperatures. Such images have wide application in energy conservation, military intelligence, and climatology and meteorology, among other fields. Ultra-violet radiation is useful for evaluating certain vegetation types and for penetrating military camouflage. Sonar uses sound waves to map the ocean floor and for coastal navigation, and radar has been employed to map areas such as parts of the Amazon Basin that are often covered with clouds and invisible to sensors in the visible and near-visible bands of the electromagnetic spectrum.

An array of satellite images has been mounted at the University of Maryland Meteorology Department web site (∼owen/EARTHCAST/). Most of these images are intended to facilitate weather and climate analysis but some may be used for viewing other aspects of the physical earth, as well. Many United States government sites also provide satellite images that are designed for instructional uses. In addition to the USGS Earthshots site already noted, the National Aeronautics and Space Administration (NASA) also provides many satellite images at

The Virtual Geography Department at, maintained by the University of Texas Department of Geography, links to an Aral Sea site that has many satellite and space shuttle images of the sea as well as surrounding areas of Central Asia. This site deals specifically with the reduction in area of the Aral Sea during the past thirty years as more and more water is taken for irrigation from the rivers that feed it. Some of the photography on this site is very impressive.


As the discussion above demonstrates, primary sources in geography cover an exceptionally diverse range of possibilities. In reality, any firsthand source that has spatial characteristics or that deals with environmental issues could be useful for helping students learn about geography. Critical thinking skills may be brought into play in the analysis of nearly all of these materials. Understanding and interpreting narrative, statistical, or visual data such as that described in this chapter requires that students be able to classify the materials, compare and contrast the critical elements in them, manipulate them, and draw conclusions about cause and effect from the data. Sequencing is called for in working with historical materials and problem-solving and decision-making are required in determining what information is relevant to a problem and in preparing reports.

With respect to the sites noted above, and the Internet in general, the usual caveats apply—sites vary greatly in quality, some data are aggregated in ways that reduce their utility for geographers, and some may contain errors or be presented in such a way as to lead to questions about their quality, lack of bias, or general usefulness as teaching/learning materials. We have tried to screen out the least useful sites and focus on ones that seem especially relevant. But the number of sites grows exponentially and there will be new areas of the Internet to explore, old sites will disappear or change their content, sites that are excellent now may become dated or lose their relevance, while sites we may have opted not to include may add important new data that increases their value. One should be constant in the search for new and better materials to use in the classroom.